SRC/clanhf.f line 975 S = S + ( AA / SCALE )**2 SCALE is 0 when N=1 Not sure if this is expected but that S (inf) is then used in the VALUE = SCALE*SQRT( S ) both of these should cause floating point exception on non-IEEE systems |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | [Lapack] xdrvrf1 doesn't protect against div by 0, Ake Sandgren |
|---|---|
| Next by Date: | [Lapack] missing dgels.f routines, Jim Brown |
| Previous by Thread: | [Lapack] xdrvrf1 doesn't protect against div by 0, Ake Sandgren |
| Next by Thread: | [Lapack] xlanhf div by zero, julie langou |
| Indexes: | [Date] [Thread] |